Chainer : a Next-Generation Open Source Framework for Deep Learning

By Seiya Tokui, Kenta Oono. 2015

812 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Model Architectures

Citations in the corpus

context2vec: Learning Generic Context Embedding with Bidirectional LSTM (2016, methodology): “context2vec We implemented our model using the Chainer toolkit (Tokui et al., 2015), and Adam (Kingma and Ba, 2014) assigned to the same batch.”
PyTorch: An Imperative Style, High-Performance Deep Learning Library (2019, methodology): “This design was pioneered for model authoring by Chainer[5] and Dynet[7].”
f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization (2016, methodology): “We use the deep learning framework Chainer [32], version 1.8.1, running on CUDA 7.5 with CuDNN v5 on NVIDIA GTX TITAN X.”
Wav2Letter++: A Fast Open-source Speech Recognition System (2018, methodology): “ESPNet uses Chainer [16] or PyTorch [17] as a back-end to train acoustic models.”
Theano: A Python framework for fast computation of mathematical expressions (2016, background): “Neon 10 and Chainer [17] are two other machine learning frameworks written in Python, with GPU kernels, that feature symbolic computation graphs and symbolic differentiation.”
Dynamic Coattention Networks For Question Answering (2016, methodology): “All models are implemented and trained with Chainer (Tokui et al., 2015).”
Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning (2017, methodology): “The Tensorﬂow implementation for the experiments is available at https://github.com/takerum/vat tf. Chainer [40] implementation is also available at https://github.com/takerum/vat chainer.”
SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size (2016, background): “, 2015a) port of SqueezeNet: (Haria, 2016) • Chainer (Tokui et al., 2015) port of SqueezeNet: (Bell, 2016) • Keras (Chollet, 2016) port of SqueezeNet: (DT42, 2016) • Torch (Collobert et al.”
Distributional Smoothing with Virtual Adversarial Training (2015, methodology): “We forgo this computation in this very paper, however, because the implementation of his procedure is not straightforward in some of the standard deeplearning frameworks (e.g. Caffe (Jia et al., 2014), Chainer (Tokui et al., 2015)).”
QSGD: Communication-Optimal Stochastic Gradient Descent, with Applications to Training Neural Networks (2016, background): “There is an extremely rich area studying algorithms and systems for efﬁcient distributed large-scale learning, e.g. [5, 10, 1, 3, 34, 28, 9, 19].”
Deep Learning for IoT Big Data and Streaming Analytics: A Survey (2017, background): “Chainer [85] is a framework that follows the latter pattern and can handle dynamic changes of the model.”
A genetic programming approach to designing convolutional neural network architectures (2017, methodology): “We have implemented our methods using the Chainer [34] (version 1.”
Graph Warp Module: an Auxiliary Module for Boosting the Power of Graph Neural Networks (2019, methodology): “We implement all models in Chainer [25] and Chainer-Chemistry 2 , an extension library of the Chainer for chemical data analysis.”
Batch Policy Gradient Methods for Improving Neural Conversation Models (2017, methodology): “Implementation Details: We implement our methods using Chainer (Tokui et al., 2015), and group sentences of the same length together in the same batch to make use of GPU parallelisation.”
ESPnet: End-to-End Speech Processing Toolkit (2018, methodology): “In the ES-Pnet, main neural network training and recognition parts are written in python, which calls Chainer and PyTorch by switching the backend option.”
Automatic differentiation in machine learning: a survey (2015, background): “In contrast, the lineage of recent frameworks led by autograd (Maclaurin, 2016), Chainer (Tokui et al., 2015), and PyTorch (Paszke et al.”
Controlling Output Length in Neural Encoder-Decoders (2016, methodology): “We use Chainer (Tokui et al., 2015) to implement our models.”